Requirements become increasingly high for random number generation as random numbers are widely applied in the computer field. Currently, there are two commonly-used random number generation methods. One is a software method in which system time is used as a seed for random number generation, and then a random number is generated based on a system function. The other is a hardware method in which a hardware device generating a random number is designed to generate a random number based on a random signal.
However, a system default random number generation method used by the software has the following disadvantage: Random numbers generated at the same time point (the same second, or the same millisecond) have a high repetition rate, resulting in imbalanced distribution of random numbers, that is, pseudorandom numbers are generated. In a computer application system, time is used as a seed for random number generation, therefore, due to the same time seeds, random numbers generated within a relatively short time such as duration including same milliseconds or seconds have a high repetition rate, resulting in imbalanced distribution of random numbers.